For more than 40 years, pregnant women around the world sought help for morning sickness through a combination of the two main ingredients in Diclegis: pyridoxine and doxylamine. An estimated 35 million are believed to have taken the medication, which is endorsed by health agencies and obstetricians worldwide. Could their faith in the treatment be based on flawed data?
That unsettling question is raised by an analysis published Wednesday in PLOS One by researchers in Toronto who reviewed more than 7,200 pages of data from a clinical trial in the 1970s that had never been made public before. The information was obtained from the FDA through the Freedom of Information Act and is part of a global initiative by the scientific community to bring to light or restore invisible and abandoned trials (RIAT). The effort began in 2013 following a call by Peter Doshi, then a postdoctoral fellow at Johns Hopkins University, and others for researchers to publish data sitting around on their computers or in file cabinets. Science magazine described it as a “stark warning” to drug companies that if they don't publish their data, others will.
Nav Persaud, a family physician and researcher at St. Michael's Hospital who is the main author of the new analysis, said the trial in question was key to obtaining the drug's approval in Canada, where it is sold as Diclectin.
The old clinical trial was conducted by Merrell-National Laboratories, which no longer exists, and involved 2,308 patients at 14 clinics in the United States who were in the first 12 weeks of pregnancy who complained of nausea or vomiting. They were randomized into eight groups that tested different medications. Women were asked to take two tablets at bedtime and, if necessary, another in the morning and midafternoon for seven nights. Doctors were asked to note the frequency of vomiting and hours of nausea.
In the medical literature over the past few decades, Persaud said that this study, which was never finished, had been referred to as being a success but with no details. However, he said, when he looked at the raw data it contained critical flaws, including the fact that some information that was supposed to be obtained during patient visits was not, the criteria for reporting improvement were unclear, outcome data was unavailable for 37 percent of participants in the placebo group and about 30 percent of the patients were lost to follow up despite the fact that the trial lasted only a week.
In an interview, Persaud called these “unusual and striking problems.”
Ron Vaillancourt, a spokesman for Duchesnay, which sells the drug in Canada (Duchesnay USA sells it in the United States), said that it is the only prescription medication for the treatment of nausea and vomiting of pregnancy approved by Health Canada and that its two active ingredients are “the most studied drug combination used in pregnancy.” He cited 16 cohort studies, two meta-analyses, an ecological study, a neurological development study and numerous others that support its use. He said that in 2016 a Health Canada advisory committee reiterated its position in a safety review.
“We have complete confidence in the safety and efficacy of Diclectin and are very proud to provide it as a safe and effective treatment option for women suffering from nausea and vomiting of pregnancy,” Vaillancourt said in a statement responding to questions from The Washington Post.
Persaud said his interest in the drug started with a pregnant patient.
“She asked whether it was a good idea to take it. I assured her it was and it was the first-line treatment. After she left I was a little bit unsettled by the conversation,” he recalled.
He went back to the guidelines and noted that they didn't cite studies but only provided a reference to a product monograph that had been provided by the pharmaceutical company. So he began digging. Persaud contacted the Canadian health agency, the company that sells the drug, and even tried to reach out to the researchers who conducted the original study. Most of them had passed away. He was surprised by the difficulty he had in getting basic information.
“I’ve been looking into this [for the] last five years and found no good reason to prescribe this medication over others. Astonishingly, I’ve found the study that is supposed to be the basis of the claim this medication is effective is problematic,” he said.
Persaud said he no longer prescribes the drug to his patients and called on the FDA, Health Canada and medical groups that issue clinical practice guidelines to conduct their own reviews. He is releasing all of the information he obtained online so that others can do their own analyses. He said there are several alternative medications for morning sickness that have undergone rigorous clinical trials that have been published with more transparency.
“Everything related to this medication should be revisited,” he said. “Until there is clear evidence this medication is effective, clinicians should stop prescribing it and the implications are that patients should stop taking it.”